Power Grid Network Evolutions for Local Energy Trading

The shift towards an energy Grid dominated by prosumers (consumers and producers of energy) will inevitably have repercussions on the distribution infrastructure. Today it is a hierarchical one designed to deliver energy from large scale facilities to end-users. Tomorrow it will be a capillary infrastructure at the medium and Low Voltage levels that will support local energy trading among prosumers. In our previous work, we analyzed the Dutch Power Grid and made an initial analysis of the economic impact topological properties have on decentralized energy trading. In this paper, we go one step further and investigate how different networks topologies and growth models facilitate the emergence of a decentralized market. In particular, we show how the connectivity plays an important role in improving the properties of reliability and path-cost reduction. From the economic point of view, we estimate how the topological evolutions facilitate local electricity distribution, taking into account the main cost ingredient required for increasing network connectivity, i.e., the price of cabling.

💡 Research Summary

The paper investigates how the distribution network—specifically the medium‑ and low‑voltage (MV/LV) layers—must evolve to support a future electricity system dominated by prosumers who both generate and consume energy locally. Building on the authors’ previous statistical analysis of the Dutch power grid, this work moves from description to design by employing Complex Network Analysis (CNA) as a synthesis tool.

First, the authors outline the socioeconomic and technological drivers behind the transition: deregulation, widespread adoption of small‑scale renewable generators (PV panels, micro‑wind, micro‑CHP), and the emergence of peer‑to‑peer (P2P) energy markets. They argue that the traditional hierarchical grid, optimized for bulk transport from large power plants, will be inadequate for the anticipated dense, localized exchanges that will occur primarily in the MV/LV segments.

The methodological core consists of generating synthetic network topologies using four well‑known growth models: Erdős‑Rényi random graphs (ER), Watts‑Strogatz small‑world graphs (WS), Barabási‑Albert scale‑free graphs (BA), and a hybrid model that combines features of the previous three. Each synthetic graph is calibrated to have the same number of nodes as the real Dutch MV/LV network, and edge weights are assigned based on actual cable characteristics (resistance, current‑carrying capacity, and unit cost).

Network performance is evaluated through a suite of CNA metrics: average degree ⟨k⟩, clustering coefficient γ, average path length L_av, characteristic path length L_cp, and betweenness centrality. The authors compare these metrics against those measured on the real Dutch grid to assess how closely each synthetic topology mimics the existing infrastructure and to identify which model best supports local energy trading.

Results show that the small‑world (WS) model consistently achieves a high clustering coefficient while maintaining short average and characteristic path lengths. This combination translates into reduced electrical losses and lower transaction distances for prosumers, making WS the most suitable candidate for a P2P‑friendly distribution network. Scale‑free (BA) graphs exhibit very short path lengths due to hub nodes, but the concentration of connectivity in a few hubs raises concerns about overload and vulnerability to targeted failures. Random (ER) graphs display low clustering and longer paths, indicating poor suitability for dense local exchanges.

To address the economic dimension, the authors incorporate cable cost data from the Northern Netherlands MV/LV networks. They calculate the incremental investment required to increase connectivity by 10 %–30 % (i.e., adding edges to raise ⟨k⟩). Although the additional cabling raises capital expenditure by roughly 5 %–12 % depending on the degree of reinforcement, the simulated reduction in average path length yields a 3 %–5 % decrease in line losses and a corresponding 4 %–6 % drop in the effective electricity price for end‑users. The net effect is a positive cost‑benefit balance, especially for the WS topology where the trade‑off is most favorable.

The paper’s novelty lies in repurposing CNA from a purely analytical framework (traditionally used for high‑voltage resilience studies) to a prescriptive design tool for distribution networks. By coupling topological metrics with realistic cost data, the authors provide a quantitative roadmap for grid operators and policymakers to plan infrastructure upgrades that enable decentralized markets.

In the discussion, the authors acknowledge limitations: the analysis does not embed detailed power‑flow simulations (voltage profiles, thermal limits) nor the ICT/control layer required for real‑time market clearing. They propose future work that integrates dynamic load models, storage technologies, and market mechanisms to validate the robustness of the suggested topologies under realistic operating conditions.

Overall, the study demonstrates that modest increases in network connectivity—particularly when guided by small‑world principles—can substantially improve reliability, reduce transmission costs, and create a more conducive environment for local energy trading, all while keeping additional cabling investments within economically acceptable bounds.